Hi,
I'm performing eigenvalue analysis on a structure present in the examples of verification (Ch4-RC-03). I think there must be some modeling error, because the periods are different from those indicated in the report of Fajfar "seismic assesment of the spear test structure." If I can get some advice as I perform pushover analysis of this structure. thanks
comparing periods of vibration
-
seismosoft
- Posts: 1263
- Joined: 06 Jul 2007, 04:55
Re: comparing periods of vibration
raf,
Are you comparing identical models, featuring the same cross-section characteristics (e.g. the ones in the SeismoStruct model feature rebars), same concrete elastic modulus (e.g. the one in SeismoStruct is being computed as 4.7xSQRT(fc)), same inertia mass, etc?
If the models feature different input assumptions, then differences in results are not necessarily unexpected, as discussed in previous posts on this Forum (search for "period").
Seismosoft Support
Are you comparing identical models, featuring the same cross-section characteristics (e.g. the ones in the SeismoStruct model feature rebars), same concrete elastic modulus (e.g. the one in SeismoStruct is being computed as 4.7xSQRT(fc)), same inertia mass, etc?
If the models feature different input assumptions, then differences in results are not necessarily unexpected, as discussed in previous posts on this Forum (search for "period").
Seismosoft Support
Re: comparing periods of vibration
Hi,
4700xSQRT(fc) MPa?
in order to make a comparison, how to calculate the total mass of the structure?
I have some problems:
-Can I apply a distribuited load on the slab that discharge in two direction?
- in the Eigenvalue analysise can I apply all the mass in a master node?
Thanks
4700xSQRT(fc) MPa?
in order to make a comparison, how to calculate the total mass of the structure?
I have some problems:
-Can I apply a distribuited load on the slab that discharge in two direction?
- in the Eigenvalue analysise can I apply all the mass in a master node?
Thanks
Re: comparing periods of vibration
Hi raf,
- Yes, i believe that is the expression.
- You can easily verify the total mass of the structure performing a static analysis and compute the mass from the total vertical reactions.
- You can apply distributed masses (mass/length) along elements - not mass/area. For instance, you can estimate an equivalent distributed mass on the beams considering the tributary area and the support conditions of the slab. If you prefer, you might also assume a series of "equivalent" beams that would simulate the slab.
- I won't recommend you to do so for 2 reasons: (1) you'll need to compute the location of the centre of mass and the associated rotational mass (something that is explicitly considered if the masses are correctly distributed over the model), and (2) that solution is more prone to, always dispensable, numerical difficulties.
- Yes, i believe that is the expression.
- You can easily verify the total mass of the structure performing a static analysis and compute the mass from the total vertical reactions.
- You can apply distributed masses (mass/length) along elements - not mass/area. For instance, you can estimate an equivalent distributed mass on the beams considering the tributary area and the support conditions of the slab. If you prefer, you might also assume a series of "equivalent" beams that would simulate the slab.
- I won't recommend you to do so for 2 reasons: (1) you'll need to compute the location of the centre of mass and the associated rotational mass (something that is explicitly considered if the masses are correctly distributed over the model), and (2) that solution is more prone to, always dispensable, numerical difficulties.
Re: comparing periods of vibration
thanks romain and seismosoft.
I'm sorry but I have not yet solved for the structure Ch-Rc-3. i modeled this structure with another program and the results are the same as those of Fajfar.
Ec=25000MPa
--------------------------------------------------
FLOOR 1&2
center of mass:
X = 4.53 m
Y = 5.29 m
Mass:
65,5t
mass moment of inertia:
1254tm^2
------------------------------------------------------
ROOF
center of mass:
X = 4.57 m
Y = 5.33 m
Mass:
64,1t
mass moment of inertia:
1196tm^2
-------------------------------------------------
so:
T1=0,559s; T2=0,476s; T3=0,385s; T4=0,195s etc
can you help me seismostruct?
I'm sorry but I have not yet solved for the structure Ch-Rc-3. i modeled this structure with another program and the results are the same as those of Fajfar.
Ec=25000MPa
--------------------------------------------------
FLOOR 1&2
center of mass:
X = 4.53 m
Y = 5.29 m
Mass:
65,5t
mass moment of inertia:
1254tm^2
------------------------------------------------------
ROOF
center of mass:
X = 4.57 m
Y = 5.33 m
Mass:
64,1t
mass moment of inertia:
1196tm^2
-------------------------------------------------
so:
T1=0,559s; T2=0,476s; T3=0,385s; T4=0,195s etc
can you help me seismostruct?
-
seismosoft
- Posts: 1263
- Joined: 06 Jul 2007, 04:55
Re: comparing periods of vibration
raf,
As you will have seen in the Verification Manual, the eigenvalue algorithm in SeismoStruct has been verified (through comparisons against literature results and other software), providing essentially identical output.
Hence, if you are obtaining different eigenvalue results when comparing SeismoStruct results with those obtained using other software packages, then this can only mean that the models that you have created are not identical amongst themselves, as we stated already in a previous response.
And, indeed, if we look at the mass information that you have provided in your message, it is immediate to see that you are comparing two different models, since the SeismoStruct model features a total mass of 212 ton, whilst your "other-software" model features a mass of 195 ton. And we suspect that these are not the only differences between the two models (mass modelling approach seems to be different as well, and we would not be surprised if the same applies to the manner in which element stiffness is being defined).
Seismosoft Support
As you will have seen in the Verification Manual, the eigenvalue algorithm in SeismoStruct has been verified (through comparisons against literature results and other software), providing essentially identical output.
Hence, if you are obtaining different eigenvalue results when comparing SeismoStruct results with those obtained using other software packages, then this can only mean that the models that you have created are not identical amongst themselves, as we stated already in a previous response.
And, indeed, if we look at the mass information that you have provided in your message, it is immediate to see that you are comparing two different models, since the SeismoStruct model features a total mass of 212 ton, whilst your "other-software" model features a mass of 195 ton. And we suspect that these are not the only differences between the two models (mass modelling approach seems to be different as well, and we would not be surprised if the same applies to the manner in which element stiffness is being defined).
Seismosoft Support